High-efficiency acceleration of an electron beam in a plasma wakefield accelerator

Journal name:
Nature
Volume:
515,
Pages:
92–95
Date published:
DOI:
doi:10.1038/nature13882
Received
Accepted
Published online

High-efficiency acceleration of charged particle beams at high gradients of energy gain per unit length is necessary to achieve an affordable and compact high-energy collider. The plasma wakefield accelerator is one concept1, 2, 3 being developed for this purpose. In plasma wakefield acceleration, a charge-density wake with high accelerating fields is driven by the passage of an ultra-relativistic bunch of charged particles (the drive bunch) through a plasma4, 5, 6. If a second bunch of relativistic electrons (the trailing bunch) with sufficient charge follows in the wake of the drive bunch at an appropriate distance, it can be efficiently accelerated to high energy. Previous experiments using just a single 42-gigaelectronvolt drive bunch have accelerated electrons with a continuous energy spectrum and a maximum energy of up to 85 gigaelectronvolts from the tail of the same bunch in less than a metre of plasma7. However, the total charge of these accelerated electrons was insufficient to extract a substantial amount of energy from the wake. Here we report high-efficiency acceleration of a discrete trailing bunch of electrons that contains sufficient charge to extract a substantial amount of energy from the high-gradient, nonlinear plasma wakefield accelerator. Specifically, we show the acceleration of about 74 picocoulombs of charge contained in the core of the trailing bunch in an accelerating gradient of about 4.4 gigavolts per metre. These core particles gain about 1.6 gigaelectronvolts of energy per particle, with a final energy spread as low as 0.7 per cent (2.0 per cent on average), and an energy-transfer efficiency from the wake to the bunch that can exceed 30 per cent (17.7 per cent on average). This acceleration of a distinct bunch of electrons containing a substantial charge and having a small energy spread with both a high accelerating gradient and a high energy-transfer efficiency represents a milestone in the development of plasma wakefield acceleration into a compact and affordable accelerator technology.

At a glance

Figures

  1. Three-dimensional particle-in-cell simulation of beam-driven plasma wakefield interaction.
    Figure 1: Three-dimensional particle-in-cell simulation of beam-driven plasma wakefield interaction.

    a, A slice through the centre of an unloaded plasma wake, where x is the dimension transverse to the motion, and ξ = z − ct is the dimension parallel to the motion, Ez is the on-axis longitudinal electric field (red solid line) and Ib is the current of the input beam (blue dotted line). b, A plasma wake generated by the same drive bunch as in a when loaded by a trailing bunch. The plasma electron density is represented in blue, while the beam density is represented in red. The ion density (not shown) is uniform. The particle-in-cell code QuickPIC9, 10 was used to generate this simulation of the beam–plasma interaction.

  2. Energetically dispersed beam profiles.
    Figure 2: Energetically dispersed beam profiles.

    a, The dispersed electron beam profile without plasma interaction, where the spectrometer is set to image 22.35 GeV. Because the beam divergence is small, the entire spectrum of the beam is well resolved even at this imaging set point. b and c, The dispersed beam profile after the electron bunches have interacted with the plasma, where the spectrometer is set to image 20.35 GeV and 22.35 GeV, respectively: Efocus in b and c. The left x and top y axes correspond to the actual scale of the electron beam recorded at the spectrometer diagnostic screen, while the bottom E axis shows the calibrated energy axis along the dispersive dimension y. The scaling factors in ac apply to the colour scale, quantifying transverse charge density. d, The spatially integrated spectrum (in x) or the linear charge density of the bunches shown in c (solid blue line) along with the final spectrum obtained from the simulation depicted in Fig. 1b (solid green line in d). The core of the accelerated trailing beam is shown for the data (dashed red line).

  3. Spatially integrated electron beam spectra from the data set.
    Figure 3: Spatially integrated electron beam spectra from the data set.

    All 92 shots from the data set are shown with the imaging spectrometer set to image 22.35 GeV. The colour scale represents the energy spectrum in nC GeV-1, similar to the blue curve of Fig. 2d, but transformed using the known dispersion of the spectrometer. The shots are sorted by the total energy-transfer efficiency (black line), as calculated using all charge above 20.35 GeV. The core energy-transfer efficiency, as calculated using only the charge found in the accelerated core of the trailing bunch, is shown as the red line. The black (or red) horizontal bar represents the typical systematic uncertainty of ±5% (or ±3%) for the total energy-transfer efficiency (or core energy-transfer efficiency).

  4. Energy-transfer efficiency dependence on wake loading.
    Figure 4: Energy-transfer efficiency dependence on wake loading.

    The total efficiency (black circles) and core efficiency (red diamonds) versus the initial ratio of the charge in the trailing bunch to that in the drive bunch before entering the plasma source. The circles (diamonds) represent the same data that make up the black curve (red curve) in Fig. 3. The black (red) vertical bar represents the typical systematic uncertainty of ±5% (±3%) for the total energy-transfer efficiency (core energy-transfer efficiency). In this data, the total charge is held constant, and only the ratio of the charge of the trailing bunch to the charge of the drive bunch is varied.

  5. FACET experimental area schematic.
    Extended Data Fig. 1: FACET experimental area schematic.

    Electron beam line features: a, beam notching device, b, transverse deflecting structure, c, initial spectrometer, d, final-focus quadrupole magnets, e, lithium plasma ionization laser, f, lithium vapour column, g, spectrometer imaging quadrupole magnets, h, spectrometer dipole magnet, and i, Cherenkov and phosphor screens. Bend dipole magnets in the ‘W’-shaped chicane are each labelled ‘D’. The arrow beneath the e symbol indicates the electron beam’s direction of motion (left to right).

  6. Measured longitudinal profile of two-bunch beam.
    Extended Data Fig. 2: Measured longitudinal profile of two-bunch beam.

    Image of a typical two-bunch beam streaked onto a profile monitor screen by the transverse deflecting radio-frequency structure (Extended Data Fig. 1b). The drive bunch appears on the right-hand side. Overlaid on the image is the projected longitudinal profile (red line). The left (x) and top (y) axes show the transverse dimensions of the streaked beam on the profile monitor screen, while the colour axis indicates the charge density of the transverse profile. The bottom (z) axis shows the streaked dimension (y) with the appropriate scaling factor applied to give the corresponding longitudinal coordinate. The right axis shows the linear charge density corresponding to the projected longitudinal profile.

  7. Lithium vapour column density profile.
    Extended Data Fig. 3: Lithium vapour column density profile.

    The profile of the neutral vapour pressure density of the lithium vapour column deduced from the measured temperature profile (temperature versus relative distance of insertion of a thermocouple probe) along the heat pipe oven is shown as the blue line. The simple fit used to describe the density profile in our model is shown as the red line.

  8. Fit to accelerated charge.
    Extended Data Fig. 4: Fit to accelerated charge.

    The blue line is the spectral projection of the same data shot shown in Fig. 2c and d. The green line is a fit to the data using a half-Gaussian tail (cyan line) to account for the diffuse, high-angular-divergence accelerated charge plus a full, asymmetric Gaussian (red) used to describe the core of the accelerated trailing bunch after subtracting the half-Gaussian tail.

Videos

  1. Simulation of beam-driven plasma wakefield acceleration
    Video 1: Simulation of beam-driven plasma wakefield acceleration
    The video shows 120 successive frames from a simulation of the electron beam‐driven plasma wakefield acceleration process using the 3D particle-in-cell code QuickPIC9,10. The input beam and plasma parameters are based on those of the experiment. The upper frame of the movie shows a slice through the center of electron beam and plasma wake structure, where x is the dimension transverse to the motion, ξ = z-ct is the dimension parallel to the motion, and Ez is the on-axis longitudinal electric field. The plasma electron density is represented in blue, while the beam density is represented in red. The ion density (now shown) is uniform. The lower frame of the video shows the evolution of the longitudinal phase space of the electron beam, using the color scale along the bottom for the beam charge density. The video depicts the beam-plasma interaction over the full length of the plasma source, including propagation before and after the 10cm plasma density ramps on either end of the 26cm flat-top density region.

References

  1. Chen, P., Dawson, J., Huff, R. & Katsouleas, T. Acceleration of electrons by the interaction of a bunched electron beam with a plasma. Phys. Rev. Lett. 54, 693696 (1985)
  2. Ruth, R., Chao, A., Morton, P. & Wilson, P. A plasma wake field accelerator. Particle Accelerators 17, 171189 (1985)
  3. Esarey, E., Schroeder, C. B. & Leemans, W. P. Physics of laser-driven plasma-based electron accelerators. Rev. Mod. Phys. 81, 12291285 (2009)
  4. Joshi, C. & Katsouleas, T. Plasma accelerators at the energy frontier and on tabletops. Phys. Today 56, 4753 (2003)
  5. Bingham, R., Mendonca, J. & Shukla, P. Plasma based charged-particle accelerators. Plasma Phys. Contr. Fusion 46, R1 (2004)
  6. Caldwell, A., Lotov, K., Pukhov, A. & Simon, F. Proton-driven plasma-wakefield acceleration. Nature Phys. 5, 363367 (2009)
  7. Blumenfeld, I. et al. Energy doubling of 42 GeV electrons in a metre-scale plasma wakefield accelerator. Nature 445, 741744 (2007)
  8. Lu, W., Huang, C., Zhou, M., Mori, W. B. & Katsouleas, T. Nonlinear theory for relativistic plasma wakefields in the blowout regime. Phys. Rev. Lett. 96, 165002 (2006)
  9. Huang, C. et al. QuickPIC: a highly efficient fully parallelized PIC code for plasma-based acceleration. J. Phys. Conf. Ser. 46, 190199 (2006)
  10. An, W., Decyk, V. K., Mori, W. B. & Antonsen, T. M., Jr An improved iteration loop for the three dimensional quasi-static particle-in-cell algorithm: QuickPIC. J. Comput. Phys. 250, 165177 (2013)
  11. Hogan, M. et al. Multi-GeV energy gain in a plasma-wakefield accelerator. Phys. Rev. Lett. 95, 054802 (2005)
  12. Muggli, P. et al. Meter-scale plasma-wakefield accelerator driven by a matched electron beam. Phys. Rev. Lett. 93, 014802 (2004)
  13. Barov, N. et al. Ultra high-gradient energy loss by a pulsed electron beam in a plasma. IEEE Proc. (2001 Particle Accelerator Conf.) 1, 126128 (2001)
  14. Katsouleas, T. Physical mechanisms in the plasma wake-field accelerator. Phys. Rev. A 33, 20562064 (1986)
  15. Tzoufras, M. et al. Beam loading in the nonlinear regime of plasma-based acceleration. Phys. Rev. Lett. 101, 145002 (2008)
  16. Rosenzweig, J. B. et al. Experimental observation of plasma wake-field acceleration. Phys. Rev. Lett. 61, 98101 (1988)
  17. Muggli, P. et al. Photo-ionized lithium source for plasma accelerator applications. IEEE Trans. (Plasma Sci.) 27, 791799 (1999)
  18. Green, S. Z. et al. Laser ionized preformed plasma at FACET. Plasma Phys. Contr. Fusion 56, 084011 (2014)
  19. Adli, E. et al. A beam driven plasma-wakefield linear collider: from Higgs factory to multi-TeV. In Electronic Proceedings of the Snowmass 2013 Community Study on the Future of High-Energy Physics http://arxiv.org/abs/1308.1145 (2013)

Download references

Author information

Affiliations

  1. SLAC National Accelerator Laboratory, Menlo Park, California 94025, USA

    • M. Litos,
    • E. Adli,
    • C. I. Clarke,
    • S. Corde,
    • J. P. Delahaye,
    • R. J. England,
    • A. S. Fisher,
    • J. Frederico,
    • S. Gessner,
    • S. Z. Green,
    • M. J. Hogan,
    • D. Walz,
    • G. White,
    • Z. Wu,
    • V. Yakimenko &
    • G. Yocky
  2. Department of Physics, University of Oslo, 0316 Oslo, Norway

    • E. Adli
  3. Department of Physics and Astronomy, University of California Los Angeles, Los Angeles, California 90095, USA

    • W. An &
    • W. B. Mori
  4. Department of Electrical Engineering, University of California Los Angeles, Los Angeles, California 90095, USA

    • C. E. Clayton,
    • C. Joshi,
    • K. A. Marsh &
    • N. Vafaei-Najafabadi
  5. Department of Engineering Physics, Tsinghua University, Beijing 100084, China

    • W. Lu
  6. Max Planck Institute for Physics, Munich 80805, Germany

    • P. Muggli

Contributions

All authors contributed extensively to the work presented in this paper.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Extended data figures and tables

Extended Data Figures

  1. Extended Data Figure 1: FACET experimental area schematic. (80 KB)

    Electron beam line features: a, beam notching device, b, transverse deflecting structure, c, initial spectrometer, d, final-focus quadrupole magnets, e, lithium plasma ionization laser, f, lithium vapour column, g, spectrometer imaging quadrupole magnets, h, spectrometer dipole magnet, and i, Cherenkov and phosphor screens. Bend dipole magnets in the ‘W’-shaped chicane are each labelled ‘D’. The arrow beneath the e symbol indicates the electron beam’s direction of motion (left to right).

  2. Extended Data Figure 2: Measured longitudinal profile of two-bunch beam. (390 KB)

    Image of a typical two-bunch beam streaked onto a profile monitor screen by the transverse deflecting radio-frequency structure (Extended Data Fig. 1b). The drive bunch appears on the right-hand side. Overlaid on the image is the projected longitudinal profile (red line). The left (x) and top (y) axes show the transverse dimensions of the streaked beam on the profile monitor screen, while the colour axis indicates the charge density of the transverse profile. The bottom (z) axis shows the streaked dimension (y) with the appropriate scaling factor applied to give the corresponding longitudinal coordinate. The right axis shows the linear charge density corresponding to the projected longitudinal profile.

  3. Extended Data Figure 3: Lithium vapour column density profile. (191 KB)

    The profile of the neutral vapour pressure density of the lithium vapour column deduced from the measured temperature profile (temperature versus relative distance of insertion of a thermocouple probe) along the heat pipe oven is shown as the blue line. The simple fit used to describe the density profile in our model is shown as the red line.

  4. Extended Data Figure 4: Fit to accelerated charge. (220 KB)

    The blue line is the spectral projection of the same data shot shown in Fig. 2c and d. The green line is a fit to the data using a half-Gaussian tail (cyan line) to account for the diffuse, high-angular-divergence accelerated charge plus a full, asymmetric Gaussian (red) used to describe the core of the accelerated trailing bunch after subtracting the half-Gaussian tail.

Supplementary information

Video

  1. Video 1: Simulation of beam-driven plasma wakefield acceleration (8.91 MB, Download)
    The video shows 120 successive frames from a simulation of the electron beam‐driven plasma wakefield acceleration process using the 3D particle-in-cell code QuickPIC9,10. The input beam and plasma parameters are based on those of the experiment. The upper frame of the movie shows a slice through the center of electron beam and plasma wake structure, where x is the dimension transverse to the motion, ξ = z-ct is the dimension parallel to the motion, and Ez is the on-axis longitudinal electric field. The plasma electron density is represented in blue, while the beam density is represented in red. The ion density (now shown) is uniform. The lower frame of the video shows the evolution of the longitudinal phase space of the electron beam, using the color scale along the bottom for the beam charge density. The video depicts the beam-plasma interaction over the full length of the plasma source, including propagation before and after the 10cm plasma density ramps on either end of the 26cm flat-top density region.

Additional data